Example of

Big Data
Analysis
Created by Puspa Cempaka Sari
Big Data Type of Data
1. Sales Data per Customer (Point of Sales)
Date of Purchase Store_Code Store_Name Member_ID Transaction_ID Category PLU Description Quantity Net_Sales Gross_Margin
01/09/2018 02XX Store A XXX00400019XXXXX 02XX-700-0102XXXX RTD TEA 1036XX PRODUCT A 2 14545.46 6449.20
01/09/2018 02XX Store A XXX00400019XXXXX 02XX-700-0102XXXX COOKIES BISCUIT 1132XX PRODUCT X 1 8454.55 2601.01
01/09/2018 02XX Store A XXX00203880XXXXX 02XX-700-0105XXXX MINERAL WATER 1206XX PRODUCT G 1 3000.00 1166.68
01/09/2018 02XX Store A XXX00203880XXXXX 02XX-700-0107XXXX JELLY & PUDDING 1231XX PRODUCT I 1 9090.91 3727.22
01/09/2018 03XX Store B XXX00400020XXXXX 03XX-700-0105XXXX RTD COFFEE 1235XX PRODUCT M 2 8545.46 2360.56
01/09/2018 03XX Store B XXX00201287XXXXX 03XX-700-0106XXXX COLD DRINKS 1263XX PRODUCT Y 2 18182.00 10255.32
01/09/2018 03XX Store B XXX00201391XXXXX 02XX-700-0108XXXX RTD TEA 4064XX PRODUCT B 2 12727.27 4810.61
01/09/2018 04XX Store C XXX00201391XXXXX 04XX-700-0104XXXX PEANUTS 4075XX PRODUCT P 1 10909.09 3409.09
01/09/2018 04XX Store C XXX00204057XXXXX 04XX-700-0109XXXX CARBONATED DRINK 9900XX PRODUCT C 1 3181.82 352.29
01/09/2018 04XX Store C XXX00204057XXXXX 04XX-700-0109XXXX FUNCTIONAL DRINK 9901XX PRODUCT J 2 6363.64 1999.98

2. Customers Profile Data

Member_ID Gender Birthday Date
XXX00400019XXXXX Female 01/10/1985
XXX00203880XXXXX Male 10/01/1990
XXX00400020XXXXX Male 03/05/1980
XXX00201287XXXXX Female 05/05/1990
XXX00201391XXXXX Male 28/10/1992
XXX00204057XXXXX Female 31/01/1988
XXX00102430XXXXX Female 15/07/1989
XXX00205690XXXXX Female 10/10/1989
XXX00202563XXXXX Male 04/01/1980
XXX00133478XXXXX Male 27/03/1982

Example of Big Data Analysis 2

Consumers Analysis
1. Identifying the consumers’ type
2. Understanding consumer’s behavior
3. Evaluate consumer’s trend

Example of Big Data Analysis 3

Customers’ Type based on The Purchase Frequency
232,389 customers 108,974 customers

Average customers
per month:
341,363

❑ From 341,363 total customers, 68% of them are 68% 32%

one-time purchasers. Purchase Only One Time Purchase Multiple Times
❑ 32% of them repeat their purchases, but only 2% Purchased only one time and did 2% of this multiple purchasers
not purchase again. purchased ≥10 times
of the total customers purchase more than 10
times.

Note: The data presented is a dummy

Example of Big Data Analysis 4

Customers’ Monthly Trend based on The Purchase Frequency
Total Purchasers One-time Purchasers Repeat Purchasers

January 2016 111,139 63% 37%

February 2016 354,221 65% 35%

March 2016 465,882 64% 36%

April 2016 433,056 62% 38%

May 2016 375,375 64% 36%

June 2016 458,285 69% 31%

July 2016 341,212 69% 31%

August 2016 280,827 79% 21%

September 2016 284,729 73% 27%

October 2016 250,336 75% 25%

November 2016 317,089 55% 45%

December 2016 424,201 79% 21%

Note: The data presented is a dummy

❑ The number of repeat purchasers in November increased almost twice compared to the previous month.
The ratio for that month is the highest in 12 months.
❑ The spike of multiple-time purchasers in November is possible due to the heavy promotion.
Example of Big Data Analysis 5
Customers’ Profile based on The Purchase Frequency

Profile By Purchase Frequency

All Purchasers One Time Purchasers Repeat Purchasers

(n= 341,363) (n= 232,389 ) (n= 108,974)
Gender
Female 53% 40% 66%

Male 47% 60% 34%

❑ One-time purchasers were
mostly dominated by male
purchasers aged around Age Group
20 – 39 years old
10 - 19 y.o 3% 3% 3%
❑ While females tend to 20 - 29 y.o 38% 34% 45%
repeat their purchases of
this product, especially 30 - 39 y.o 32% 36% 28%

the young adult group (20- 40 - 49 y.o 21% 20% 17%

29 years old) 50 - 59 y.o 5% 6% 5%

60 y.o and Older 1% 1% 1%

Note: The data presented is a dummy

Example of Big Data Analysis 6

Customers’ Type based on The Product Purchase
Product A
Average total purchasers per month: Product B

341,363

24% 22% 41%

❑ Product B has the highest exclusive
purchasers (41%) among all the
products
1%
❑ Among multiple-product
purchasers, customers tend to buy
products A and B together than 5% 4%
other combination

Note: The data presented is a dummy

3%
Product Purchasers % of Purchasers
Product A Purchasers 68%
Product B Purchasers 52%
Product C Purchasers 13%

Product C
Example of Big Data Analysis 7
Customers’ Profile based on The Product Purchase
Product A Purchasers Product B Purchasers Product C Purchasers Three Product
(Exclusive) (Exclusive) (Exclusive) Purchasers
(n=81,927) (n=139,959) (n=10,241) (n=3,414)

Gender

Female 71% 67% 51% 47%

❑ Product A and B has Male 29% 33% 49% 53%

more female purchasers
than male purchasers
❑ While Product C is more Age Group
equally split between
10 - 19 y.o 3% 3% 5% 3%
genders, but heavier on
males compared to 35% 36% 40% 42%
20 - 29 y.o
other product
30 - 39 y.o 35% 37% 34% 34%
❑ Product C is slightly
more popular among 40 - 49 y.o 21% 17% 15% 16%
20-29 y.o compared to
other product 50 - 59 y.o 5% 6% 5% 4%

60 y.o and Older 1% 1% 1% 1%

Note: The data presented is a dummy

Example of Big Data Analysis 8

RFM Analysis
❖ RFM stands for :
✓ Recency - How recently did the customer purchase?
✓ Frequency - How often do they purchase?
✓ Monetary Value - How much do they spend?

❖ Basically, RFM analysis is a marketing technique, used to

determine quantitatively which customers are the best ones
by examining how recently a customer has purchased
(Recency), how often they purchase (Frequency), and how
much the customer spends (Monetary).

❖ RFM analysis created based on the marketing axiom that

"80% of your business comes from 20% of your customers.“

❖ Generally, the purchasers who buy more recently, more

frequently with higher monetary are seen as promising
purchasers.

Example of Big Data Analysis 9

RFM Analysis of Product A
Percentage of Percentage of
No RFM Group Number of Purchasers Total Sales
Purchasers Sales
❑ RFM analysis can be used to
1 R1F1M1 6,735 76,786,314 10% 30% evaluate the contribution for
each purchaser's segment to
2 R2F1M1 4,513 26,702,422 4% 7% the business.
3 R1F2M1 5,072 22,511,509 6% 6%
❑ In product A case, the biggest
4 R2F2M1 6,197 34,753,946 9% 10% contribution to the total sales
comes from the R1F1M1
5 R1F2M2 15,778 69,848,240 35% 23%
group. With the 10% of
6 R2F2M2 16,649 74,564,153 37% 24% purchasers, they contribute
Total 54,944 305,166,584 100% 100%
30% to the total sales.
❑ If it is compared to other
Note : R1F1M1 Recent, Purchase more than one time with the largest monetary product purchasers, the total
Recency : R1 ---> 21 Aug - 23 Oct R2F1M1 Not Recent, Purchase more than one time with the largest monetary contribution from product A
R2 ---> 1 Jun - 20 Aug good purchasers (group 1-4) to
R1F2M1 Recent, Purchase only one time with the largest monetary the total sales (53%) is smaller
Frequency : F1 ---> more than 1 time
R2F2M1 Not Recent, Purchase only one time with the largest monetary than other products (product
F2 ---> 1 time
R1F2M2 Recent, Purchase only one time with the smallest monetary
B = 76% and C = 80%).
Monetary : M1 ---> 5,500 and more
M2 ---> less than 5,500 R2F2M2 Not Recent, Purchase only one time with the smallest monetary

Note: The data presented is a dummy

Example of Big Data Analysis 10

RFM Analysis of Product A : Promising VS Non-Promising Purchasers
❑ If promising purchasers are the purchasers Promising Purchasers Non-Promising
who give the biggest contribution, then (n=6,735) Purchasers (n=16,649)
non-promising purchasers are the
purchasers who give the smallest Gender
contribution to the total sales (R2F2M2). Female 43% 53%
Male 57% 47%
❑ The profile of promising and non-
promising purchasers for product A is Age Group
different. If the promising purchasers are
male, aged 30 – 49 y.o, the profile of non- 10-19 y.o 3% 5%
promising purchasers is the same with 20-29 y.o 35% 41%
other promising purchasers, which are
30-39 y.o 36% 33%
female, aged 10 - 29 y.o.
40-49 y.o 20% 16%
50-59 y.o 4% 4%
❑ There is indication that non-promising
60 y.o and older 1% 1%
purchasers of product A is the other
product purchasers who tried product A
but they didn’t like it.are Note: The data presented is a dummy

Example of Big Data Analysis 11

Other Example of RFM Analysis

Category Purchasers Avg. Transaction* Avg. Quantity* Avg. Sales* Total Sales
HF - HM 2,524 2.7 1.7 50,462 127,366,088
LF - HM 2,290 1.5 2.8 50,980 116,744,200
LF - LM 3,738 1.5 1.2 50,535 188,899,830
Grand Total 8,552 1.9 1.9 50,659 433,010,118

Note: The data presented is a dummy

Note :
HF – HM = High Frequency, High Monetary (pack)
LF – HM = Low Frequency, High Monetary (pack)
LF – LM = Low Frequency, Low Monetary (pack)

* The central tendency using median score instead of the mean score, due to the non-normality data
distribution

Example of Big Data Analysis 12

Switching Behavior between the Product Purchasers
The number of existing purchasers during June – August 2018 : 350,554

Product A Product B Product C Product D Product E

Purchased during June – August 2017 292,859 287,431 251,041 237,782 223,632
Ever Purchased in Mar – May 2017 166,428 162,795 141,612 112,191 100,655
Keep purchase in June – August 2017 65% 60% 27% 45% 50%
Stop purchase in June – August 2017 35% 40% 73% 55% 50%
Stop Purchase and Switch to :
Product A 58% 56% 58% 55%
Product B 55% 49% 55% 59%
Product C 20% 19% 25% 26%
Product D 33% 32% 37% 32%
Product E 25% 26% 21% 27%

❑ Among the existing purchasers, product A and B have the highest purchasers during June – Aug than other products
❑ Switching behavior indeed happened during the P6M period (Mar-Aug). Product A, B and D tend to be chosen when the purchasers decide
to switch
❑ However, Product A and B have more loyal purchasers than other flavor, while product C purchasers seem to be the least loyal

Note: The data presented is a dummy

Example of Big Data Analysis 13

Purchasers Characteristics Note: The data presented is a dummy

CATEGORY ABC PURCHASERS PRODUCT B PURCHASERS

Product C Product D Product E2

Product B Growth Rate:
❑ ABC Category : -41%
51,193
Product A ❑ Product A : -41% 12,059 Growth Rate:
❑ Product B : -44%
❑ ED Category : -41%
❑ Product C : -37%
63,385 ❑ Product D : -42%
❑ Product E : -40%
23,162 39,518
20,400 4,517 4,202
16,464 25,847 14,602
75,254 20,437 2,272 2,732
14,083 14,599 13,020 14,662
26,575 33,906 19,857 8,458 8,702
18,154
Jan-Mar'19 Apr-Jun'19 Jul-Sep'19 Oct-Dec'19 Jan-Mar'20 Jan - Feb'19 Mar - May'19 Jun - Sep'19 Oct - Dec'19 Jan-Mar'20

ABC Purchasers Characteristic Original Tea Purchasers Characteristic

Among who still purchased the product: Among who still purchased the product:
Keep purchase ABC Category Keep purchase DE Category
Switch to Product F Switch to Product F 11% Stop buy
Stop buy 15% Switch to Product G Switch to ABC Category
the Product the Product
Switch to DE Category Switch to Product G
in next 8% in next
period 75% 5% period 73%
3% 5%
❖ Loyalty level towards ABC and ED category relatively low, as majority of its 4%
2%
purchasers were lapsed users

Example of Big Data Analysis 14

 Promotion participant (n=50,253)
Purchasers
Characteristics PROFILING : AGE
GENDER
❑ This promotion succeeded in Below 10 y.o 1%
attracting new members to 10 - 19 y.o 20%
purchase in store A during the
campaign period. 53% 20 - 29 y.o 40%

30 - 39 y.o 28%
❑ Majority of promotion
participants were either Male or 40 - 49 y.o 9%
20 – 29 y.o Males 47%
50 y.o and older 2%
Females
❑ They repeated their purchase 3
times per month and spent ±
IDR 45,000 per transaction
PURCHASE BEHAVIOR : PURCHASERS’
❑ Majority of the promo COMPOSITION
❖ Purchase 3 times/month
participants were new
purchasers or they started to ❖ Spend IDR ± 45,000/transaction
purchase in December 2015 or 55%
January 2016. Note: The data presented is a dummy
(n=24,121) (n=26,132)
Original Purchaser 45%
New Purchaser

Original Purchaser consist of :

Non Regular Purchaser 35%

Regular Purchaser 10%

Example of Big Data Analysis 15
 Product Evaluation
Analyze the performance of each of the products and identify the biggest
contribution to the sales

Example of Big Data Analysis 16

Product Performance : Before VS During the Promotion
Before Promotion During Promotion
Original Purchaser (count) Promo Participant (count) Differences
(𝐷𝑢𝑟𝑖𝑛𝑔 − 𝐵𝑒𝑓𝑜𝑟𝑒)
Average per day Average per day 𝐵𝑒𝑓𝑜𝑟𝑒
❑ Product A and Product C were the
Product A 74 134 82% most purchased G1 categories during
Product B 67%
the promotion.
60 100
Product C 58 98 69% ❑ During promotion, the number of G1
purchasers for each category
Product D 55 89 62% increased.
Product E 48 67 40% ❑ The highest increases are shown in
Product A, Product C and Product D.
Product F 44 59 34%

Product G 40 53 33%

Product H 37 49 34% Note: The data presented is a dummy

Product I 33 46 39%

Product J 30 45 50%

Product K 20 24 20%

Product L 20 23 15%

Others Food 12 15 22%

Others Beverages 11 15 33%

Product B will not be highlighted in this report because:

1. It is bought concurrently with Product A because of the promotional set nature.
2. It does not have the unique/originality of store. Example of Big Data Analysis 17
Most Purchased Variant

P R O D U C T A P U R C H A S E R S (n=5,680)
Category Number Purchaser Repeat Purchaser Repeat Frequency

Variant A1 43% 47% 4.2 times

Variant A2 42% 48% 4.1 times ❖ Variant A1 and Variant A2 were the
Variant A3 26% 37% 4.0 times most purchased product by Product
4.0 times
A purchasers during the campaign.
Variant A4 23% 36%
Variant A5 23% 33% 3.9 times ❖ Those two products also showed the
3.9 times
highest number of repeat purchaser
Variant A6 23% 35%
(above 40%).
Variant A7 23% 37% 3.8 times
3.6 times
❖ More than 30% of Variant A1 and
Variant A8 22% 35%
Variant A2 purchaser repeated their
Variant A9 22% 36% 4.1 times purchase approximately 4 times
Variant A10 20% 39% 3.9 times during the campaign.
Variant A11 17% 39% 4.0 times
4.1 times Note: The data presented is a dummy
Variant A12 17% 31%
Variant A13 11% 31% 3.6 times

Variant A14 11% 33% 3.7 times

Average item purchased : 3,53

Example of Big Data Analysis 18

Identify the Most Purchased Variant

PRODUCT C PURCHASERS (n=2,500)

Category Number Purchaser Repeat Purchaser Repeat Frequency

Variant C1 60% 55% 4.3 times ❖ Among product C, Variant C1 became

the most purchased variant during
Variant C2 56% 52% 3.6 times
the promotion.
Variant C3 47% 50% 5.2 times
❖ Variant C3 and C5 (hot & ice) had
Variant C4 35% 45% 2.8 times fewer purchasers, but it had the
highest repeat frequency among
Variant C5 33% 43% 5.7 times other variants.
Variant C6 28% 38% 2.2 times
Note: The data presented is a dummy
Variant C7 27% 35% 2.5 times

Variant C8 25% 29% 3.5 times

Variant C9 25% 27% 2.3 times

Variant C10 20% 29% 2.4 times

Variant C11 20% 26% 2.0 times

Variant C12 18% 23% 3.1 times

Variant C13 18% 21% 2.3 times

Variant C14 15% 18% 2.0 times

Variant C15 13% 15% 2.5 times Example of Big Data Analysis 19
Customers Basket Analysis
Among product A, purchasers also purchased : Objective:
Product M 75% Identifying product groups and items that
Product B 74% are most likely to be bought together, by
Product N 46% evaluating big data sets, such as purchase
Product O 39% history.
Product H 37%
Product D 37%
Product P 36%
Including the promotional set
package Product C 35% ❑ Most Product A purchasers also
Product Q 33% purchased Product M or Product B
Product G 32% along with Product A.
❑ The promotional package food might
cause this tendency during promotion.
Among Product C, purchasers also purchased : ❑ While for Product C, the purchasers
were more likely to purchase it with
Product A 54% Product A (54%) or Product D (43%).
Product D 43%
Product O 35%
Product Q 35% Note: The data presented is a dummy
Product B 35%
Product G 33%
Product H 33%

Example of Big Data Analysis 20

Customers Basket Analysis
Among purchasers of variant C1, they also purchased :
Variant D1 48% Objective:
Variant D2 37%
Variant D3 33%
Identifying product groups and items that
Variant D4 28%
are most likely to be bought together, by
Variant D5 26% evaluating big data sets, such as purchase
Variant Q1 25% history.
Variant D6 20%
Variant A1 17%
Product N1 17%
(n=1,500) Variant D7 10%
❑ Most of Variant C1 purchasers buy it
together with variants of Product D.
❑ for Variant C5 purchasers, they prefer
to purchase it with other Product C or
Among purchasers of variant C5, they also purchased : Product D variants.

Variant C1 53%
Variant D1 45% Note: The data presented is a dummy
Variant D2 40%
Variant C2 38%
Variant D3 35%
Product N1 28%
Variant A2 23%
Variant C3 20%
(n=1,578) Variant Q1 15%
Variant D6 15%

Example of Big Data Analysis 21

 Promotion Effectiveness
Analyze the key success and barrier for each promotion to re-formulate a
better and more successful promotion in the future

Example of Big Data Analysis 22

Differences Before VS During Promotion
Before Campaign During Promotion
Original Purchaser Promo XXX Participant
❑ Number of G1 purchasers per day
during the Promo XXX increased.

200 purchaser/
day 248 purchaser/
day
❑ Even though the number of G1
transaction during the promotion

125 187
Purchased Purchased slightly decreased, but the proportion
Total
Purchaser G1/day G1/day is increase.
(63%) (75%)
❑ The total products purchased during
promotion, including total G1 items
purchased per day also increased.

900 transaction/
day 587 transaction/
day ※Concurrently, total G2 products
also increased.
Total
Transaction 480 Have G1 in
the list 450 Have G1 in
the list
(53%) (77%)
Note: The data presented is a dummy

2290 product/
day 3500 product/
day

800 1000
Total Product G1 product/ G1 product/
Purchased day day
(35%) (29%)

Example of Big Data Analysis 23

Consumers Type during the Promotion
Total Purchasers
During Promotion Period

542,629
Existing Consumers
New Consumers
(Start purchase Ichitan before promotion period)
(Start purchase in the promotion period)

79% 429,058 113,571 21%

54% 25% 12% 9%

292,987 136,071 63,974 49,597

Repeat Purchaser One Time Purchasers Repeat Purchaser

One Time Purchasers
34% 20% 15% 10% 2% 10% 4% 5%

184,494 108,493 81,394 54,677 10,853 53,121 21,705 27,892

Exclusive purchaser Multi-variant Exclusive purchaser Multi-variant Exclusive purchaser Multi-variant Exclusive purchaser Multi-variant
to one variant purchasers to one variant purchasers to one variant purchasers to one variant purchasers

❑ This promotion generated 542,629 purchasers, of which 79% of them are first-time purchasers and 21% are existing consumers.

❑ 68% of the new consumers, however, made only one-time purchases, on the contrary, 34% (25% + 9%) made repeat purchases.

❑ New purchasers tend to be the exclusive purchasers of one variant while the existing purchasers use this promotion to buy multi-variant products

Note: The data presented is a dummy

Example of Big Data Analysis 24
 The Trend of New and Existing Consumers during the Promotion
New VS Existing Consumers (monthly average) Existing Consumers Come from :
New Consumers
Existing Consumers
P1M purchasers

>P1M purchasers
135,505

111,119
62,651

56,474 81,582

134,625 77,024 87%

104,524 32,823 39,568 92,747
11,448 69,281 64,044
31,701 33,354 13%
16,333 11,920
Jan - Mar'18 Apr - Jul'18 Aug - Sep'18 Oct - Dec'18 Jan - Feb'19 Mar - May'19 Jun - Sep'19 Oct - Nov'19 Jan - Nov'19 (average)

Promo A Promo B

Growth Rate Promo A VS Promo B

❑ The incremental of purchaser’s number during promotion B was mostly generated by existing consumers, with
Total +62% 87% of these existing consumers being non-recent buyers (last purchase more than 1 month)
New Consumers +34%
❑ This may indicate that JJG able to re-activate more relapse consumers
Existing Consumers +97%

Note: The data presented is a dummy

Example of Big Data Analysis 25

 The Trend of Consumers Repeat Rate during the Promotion
100% Promo A Promo B 15
90% 85%
79% 13
80% 76% 75% 75% 77%
73% 71% 70% 73%
69% 11
70% 65%
59% 56% 9
60%
9.3
50% 7.8 7
7.4 7.2
40% 6.7 6.7 6.7 6.5 6.3 6.5 5
5.6 5.9
30% 4.7 4.2 3
20%
2.6 2.6 2.3 2.3 2.3 2.5 2.5 2.6 2.6 2.6 2.5 2.8 2.5
10% 2.2 1
0% -1
Oct'18 Nov'18 Dec'18 Jan'19 Feb'19 Mar'19 Apr'19 May'19 Jun'19 Jul'19 Aug'19 Sep'19 Oct'19 Nov'19
Repeat Rate Purchase Frequency Purchase Quantity

Average: Oct – Dec’18 Jan – Feb’19 Mar – May’19 Jun – Sep’19 Oct – Nov’19
❑ Compared to promo A, the repeat rate during the promo B
(Promo A) (Promo B)
period is slightly higher.
Repeat Rate 65% 72% 71% 77% 75%
❑ Promo B also generated higher purchase frequency and
Purchase Frequency 6.9 4.8 6.4 7.4 7.5 quantity during its promotion period.
Purchase Quantity 2.5 2.3 2.5 2.6 2.7 ❑ However, there is an indication that promo B already
created less excitement in the 2nd month of the campaign
Note: The data presented is a dummy
(given the decline of purchase freq. & basket size in Nov’19)

Example of Big Data Analysis 26

Effect of the Campaign to the New Purchasers

PROFILING :
(n=8,309) ❖ Majority of the promo participants
were the new purchaser (purchased
GENDER AGE since December 2015).
Below 10 y.o 1% ❖ Majority of the new purchasers were
10 - 19 y.o 15%
females, 20-29 y.o.
Males 46% ❖ 45% of them repeat their purchase,
20 - 29 y.o 45%
Females and 15% repeat at least once a week
30 - 39 y.o 22% for the rest of the campaign.
54%
40 - 49 y.o 12%
❖ They spent ± IDR 56,000 and
50 and older 5% purchase ± 8 items per transaction.

Note: The data presented is a dummy

PURCHASE BEHAVIOR:
❖ Purchase 3 times/month.
❖ 45% of them (3,739/6,309) repeat their purchase, and 15%
(1,246/8,309) repeat at least once a week for the rest of the
campaign.
❖ Spend IDR ± 56,000/transaction.
❖ Purchased 8-9 different items/transaction.

Example of Big Data Analysis 27

Effect of the Campaign to the Original Purchasers
Non Regular Purchasers
Purchase less than 4
times a month Among Non Regular Purchaser :
❑ The proportion of one time
purchasers decreased among
BEFORE CAMPAIGN (n=15,780/88%) : CAMPAIGN PARTICIPANT (n=5,698/74%) :
campaign participants.
❖ 57% (=8,995/15,780) did not repeat their purchase ❖ 35% (=1,994/5,698) did not repeat their purchase
❖ Purchase 1-2 times/month ❖ Purchase 3-4 times/month ❑ They spent more money and bought
❖ Spend IDR ± 28,000/transaction ❖ Spend IDR ± 49,000/transaction more product during the campaign.
❖ Purchased ± 3 different items/transaction ❖ Purchased ± 8 different items/transaction

Among Regular Purchaser :

❑ Their purchase frequency decreased
Regular Purchasers
during the campaign
Purchase minimum
once a week
❑ However, their spending and
amount of products purchased
BEFORE CAMPAIGN (n=2.152/12%) : CAMPAIGN PARTICIPANT (n=2,002 /26%) :
during the campaign increased.
❖ Purchase 10 times/month ❖ Purchase 5 times/month
❖ Spend IDR ± 35,000/transaction ❖ Spend IDR ± 45,000/transaction Note: The data presented is a dummy
❖ Purchased ± 4 different items/transaction ❖ Purchased ± 7 different items/transaction

Note :
88% = 15,780/17,932; 12% = 2,152/17,932
74% = 5,698/7,700; 26% = 2,002/7,700
Example of Big Data Analysis 28
Most Contributing Purchasers to the Promotion
Promotion Cost Promotion sales
IDR 530,992,668 ± 66% IDR 350,717,680 ❑ During the campaign, the total sales
covered ± 66% of the promotion cost.
❑ The biggest sales comes from original
Original VS New Purchasers purchasers, which covered ± 35% of
the cost.
Contribution/
Type of Purchaser Total Purchaser Sales Contributed Contribution
purchasers ❑ Although the non regular purchasers
gave more contribution (total) to the
Original Purchaser 48% IDR 184,949,865 35% 0.005%
sales, but regular purchasers shows
New Purchaser 52% IDR 165,767,815 31% 0.004% higher contribution per purchasers.
This type of purchasers are small but
spent more money during the
campaign.
Non Regular VS Regular Purchasers
Note: The data presented is a dummy

Contribution/
Type of Purchaser Total Purchaser Sales Contributed Contribution
purchasers

Non Regular Purchaser 83% IDR 196,989,578 37% 0.006%

Regular Purchaser 17% IDR 153,728,102 29% 0.015%

Example of Big Data Analysis 29

Most Contributing Purchasers to the Promotion
Promotion Cost Promotion sales
IDR 530,992,668 ± 66% IDR 350,717,680

Contribution/
Type of Purchaser Total Purchaser Sales Contributed Contribution
purchasers ❑ The previous slide show that original
purchaser gives more contribution
Original Purchaser 48% IDR 184,949,865 35% 0.005%
than new purchaser. It also shows that
Non Regular Purchaser 33% IDR 98,627,870 19% 0.005%
the regular purchasers gives more
contribution to the sales.
Regular Purchaser 15% IDR 86,321,995 16% 0.010%
❑ However, the cross analysis between
New Purchaser 52% IDR 165,767,815 31% 0.004% types of customers shows the new
and regular purchasers give more
Non Regular Purchaser 50% IDR 97,484,700 18% 0.004% contribution to the sales than other
type of purchasers
Regular Purchaser 2% IDR 68,283,115 13% 0.022%

Note: The data presented is a dummy

Example of Big Data Analysis 30

Starting VS Ending of the Campaign
The beginning of The end of
the campaign the campaign
Note ❑ More purchasers came at the beginning of campaign
period.
All Purchasers
❑ More transactions also happened during the
❖ Number of purchasers 8,545 purchasers 6,892 purchasers beginning of the campaign period.
❖ Number of transaction 16,778 transaction 12,690 transaction ❑ However, more product purchased and money spent
❖ Number of different Items purchased 5.87 item/transaction 6.25 item/transaction at end of the campaign.
❖ Amount of purchase ± IDR 35,000/transaction ± IDR 48,000/transaction ❑ This shows Ponta purchasers were more excited to
participate on the beginning of campaign. However,
the customers in the end of promotion period were
Original Purchasers quality customers (who buy more and spend more).
❖ Number of purchasers 3,879 purchasers 3,030 purchasers
❖ Number of transaction 8,250 transaction 6,757 transaction
❖ Number of different Items purchased 5.78 item/transaction 5.90 item/transaction
❖ Amount of purchase ± IDR 35,000/transaction ± IDR 38,000/transaction

New Purchasers
❖ Number of purchasers 3,560 purchasers 2,589 purchasers
Note: The data presented is a dummy
❖ Number of transaction 5,680 transaction 5,020 transaction
❖ Number of different Items purchased 6.25 item/transaction 6.89 item/transaction
❖ Amount of purchase ± IDR 35,000/transaction ± IDR 45,000/transaction

Example of Big Data Analysis 31

Promotion Cost and Efficiency Calculation
Type of Promotion
Month
Promo A Promo B Promo C Promo D Promo E
(Aug 2017) (Sep 2017) (Oct 2017) (Nov 2017) (Apr - Jul 2018)
Total Cost (IDR) xxxxxx xxxxxx xxxxxx xxxxxx xxxxx
Generate New Purchasers 38,459 purchasers 65,959 purchasers 43,672 purchasers 300,126 purchasers 500,211 purchasers

Generate Repeaters (New + Existing) 7,095 purchasers 18,029 purchasers 9,023 purchasers 60,232 purchasers 400,868 purchasers

A : 10,070 purchasers A : 25,248 purchasers A : 22,562 purchasers A: 87,958 purchasers A : 150,029 purchasers
B : 8,527 purchasers B : 20,086 purchasers B : 15,504 purchasers B : 70,790 purchasers B : 135,346 purchasers
Generate Exclusive Purchasers C : 8,026 purchasers C : 5,086 purchasers C : 5,121 purchasers C : 45,988 purchasers C : 55,354 purchasers
D : 8,635 purchasers
E : 9,787 purchasers

Cost per New Purchasers 19,447 200,529 10,149 28,771 70,499

Cost per Repeaters 100,453 950,873 45,864 100,849 75,161
A : 64,744 A : 500,584 A : 16,982 A: 60,976 A : 200,948
B : 86,072 B : 700,883 B : 25,812 B : 75,871 B : 2501,065
Cost per Exclusive Purchasers C : 92,397 C : 3,499,132 C : 100,705 C : 120,201 C : 580,736
D : 7,043,011
E : 5,952,465

*Multiply by the sales percentage: 72.3% Note: The data presented is a dummy
❑ Both purchasers and repeaters were generated at the highest level in Nov (Promo D) and Apr(Promo E).
❑ In regard to the efficiency (per cost), however, neither of them (Promo D and E) were way too expensive. The new purchaser per cost for Nov was Rp.
28,771 and Promo E was Rp. 70,499. Repeater per cost for Nov was Rp. 100,849 and NWCP was Rp. 75,161. This means the newly acquired repeater for
Promo E must buy a minimum xx bottles, to break even. Example of Big Data Analysis 32
Purchaser’s Retention Rate
Purchasers Retention Rate measure the number of purchasers who
keep purchase the product until the end of specific period

(#𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑟𝑠 𝑖𝑛 𝑡ℎ𝑒 𝑒𝑛𝑑 𝑜𝑓 𝑝𝑒𝑟𝑖𝑜𝑑 − #𝑛𝑒𝑤 𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑟𝑠 𝑎𝑐𝑞𝑢𝑖𝑟𝑒𝑑 𝑑𝑢𝑟𝑖𝑛𝑔 𝑡ℎ𝑎𝑡 𝑝𝑒𝑟𝑖𝑜𝑑)
Purchasers Retention Rate = #𝑝𝑢𝑟𝑐ℎ𝑎𝑠𝑒𝑟𝑠 𝑎𝑡 𝑡ℎ𝑒 𝑠𝑡𝑎𝑟𝑡 𝑜𝑓 𝑡ℎ𝑎𝑡 𝑝𝑒𝑟𝑖𝑜𝑑

Total LY HL YY TM TH
Total Purchasers in Apr – May 415,976 245,914 231,524 123,159 47,130 21,236
Total Purchasers in June 181,217 92,048 88,812 44,434 52,718 41,356
New purchasers in June 97,320 61,873 60,728 34,077 49,072 41,140
Purchasers Retention Rate in June 20% 12% 12% 8% 8% 1%

New Purchasers in Apr – May 316,650 201,641 193,101 100,071 27,130 1,236
June Purchasers comes from new purchasers in Apr-May 58,924 30,027 29,599 17,150 3,646 216
New Purchasers Retention Ratio in June 19% 15% 15% 17% 13% 17%

❑ The higher retention rate of Promo A purchasers in June indicated that the enthusiasm to purchase the product was still high during the
campaign period, however, the enthusiasm didn’t last long, and it ended by the time the promotion was over.
Note: The data presented is a dummy
Example of Big Data Analysis 33
 Store Evaluation
Evaluate the performances of each store to identify which store has the
highest contribution and which store needs improvement also analyze top
product to sell in each type of store

Example of Big Data Analysis 34

Store Contribution to the Sales
Total Sales from 38 stores
IDR 18,464,059,748

Store Code Total Sales Ratio Cum. ratio Store Code Total Sales Ratio Cum. ratio
Note: The data presented is a dummy
Store A 1,578,348,796 9% 9% Store T 385,344,200 2% 69%
Store B 1,340,527,778 7% 16% Store U 380,279,008 2% 71% The 15th Top contributor covers 60%
Store C 879,351,221 5% 21% Store V 364,478,903 2% 73% of the total sales, in which Store A,
Store D 815,224,579 4% 25% Store W 360,800,765 2% 75% Store B and Store C have the highest
contribution to the total sales.
Store E 679,192,344 4% 29% Store X 359,980,734 2% 77%
Store F 650,500,487 4% 32% Store Y 352,098,567 2% 79%
Store G 589,358,897 3% 35% Store Z 348,976,300 2% 81%
Store H 578,445,467 3% 39% Store AA 345,098,762 2% 83%
Store I 548,487,945 3% 41% Store AB 342,087,634 2% 84%
Store J 539,028,890 3% 44% Store AC 340,892,385 2% 86%
Store K 535,722,678 3% 47% Store AD 320,784,972 2% 88%
Store L 510,685,890 3% 50% Store AE 318,098,356 2% 90%
Store M 497,457,992 3% 53% Store AF 309,460,987 2% 91%
Store N 456,489,743 2% 55% Store AG 305,089,734 2% 93%
Store O 450,424,200 2% 58% Store AH 300,478,902 2% 95%
Store P 445,235,921 2% 60% Store AI 285,908,457 2% 96%
Store Q 435,724,792 2% 62% Store AJ 263,097,245 1% 98%
Store R 420,578,934 2% 65% Store AK 230,897,654 1% 99%
Store S 410,340,975 2% 67% Store AL 189,078,654 1% 100%

Example of Big Data Analysis 35

Store Classification
The 38 stores will be categorized as good-performing stores and bad-performing stores. The indicators to be categorized are :
1. Contribution of stores in total sales during the campaign period (top 15 stores)
2. The average sales of members during the campaign period.
3. The average transaction of members during the campaign.
4. The average purchase of member Note: The data presented is a dummy

Indicators 3 Indicators 3
Indicators 1 Indicators 2
(Member average (average purchase per Name of Stores
(the most contributed) (Member average sales)
transaction) Member)
The top 15 contributing
Increased Increased Increased A57, YA, PI, MB, TA
stores
The top 15 contributing
Increased Increased Decreased K39, K42, CI, NN, AP, KW, RH, AC
stores
The top 15 contributing
Increased Decreased Increased AR
stores
The top 15 contributing
Increased Decreased Decreased SA
stores
Not the top 15 RO, AT, GM, BN, TM, PG, PA, K72, AS, BX, BR, K14,
Increased Increased Increased
contributing stores CL
Not the top 15
Increased Increased Decreased MA, MD, IR, SN, TR, NE, VA
contributing stores
Not the top 15
Decreased Increased Decreased BM, SM, GN
contributing stores

Example of Big Data Analysis 36

TOP 10 Most Purchased Product in TOP 3 Stores
Total Purchasers Total Transaction Total Sales (IDR)

Product 1 5,370 21,110 2,112,578,956

Product 2 5,300 20,900 2,091,547,896

Product 3 3,732 16,196 1,773,056,324

Product 4 3,720 14,160 723,744,533

Product 5 3,644 13,932 714,696,532

Product 6 2,132 9,396 2,820,376,655

Product 7 2,080 6,160 2,465,897,839

Product 8 1,074 4,148 881,938,740

Product 9 1,033 3,066 259,983,190

Product 10 1,030 3,060 121,496,581

❑ Among the TOP 3 Stores (based on sales contribution), Product 1 and 2 has the highest number of purchasers
❑ Despite having a smaller number of purchasers and transaction, Product 6 and 7 has the highest sales due to their higher price rate
Note: The data presented is a dummy

Example of Big Data Analysis 37

TOP 10 Most Purchased Product in TOP 3 Stores
TOP 3 Stores Store A Store B Store C
Total Purchasers Total Purchasers Total Purchasers Total Purchasers

Product 1 5,370 2,685 1,880 806

Product 2 5,300 1,855 2,650 795
Product 3 3,732 1,306 1,866 560
Product 4 3,720 1,302 558 1,860
Product 5 3,644 1,275 547 1,822
Product 6 2,132 746 320 1,066
Product 7 2,080 1,040 728 312
Product 8 1,074 537 376 161
Product 9 1,033 517 362 155
Product 10 1,030 515 361 155

❑ Product 1, 2 and 3 purchasers mostly came from stores A and B.

❑ While product 4, 5 dan 6 mostly came from store C.

Note: The data presented is a dummy

Example of Big Data Analysis 38